I have some time to myself this morning as electricians upgrade the service at the house. This had me thinking about off-grid power and I spent a couple hours diving deep into making something convenient and durable for monitoring, controlling and integrating various power supplies and consumers. I was focused on Bluetti, because that's what I have to play with and really rooting-around in the Bluetooth Low Energy (BLE) interface.
At first it looked promising. Building off the Python-based setup I experimented with earlier and some Go-based BLE experiments this morning, I was hopeful that I might be able to pull something together and started digging back into the repositories and forum posts where I found the original Python implementation.
There was a lot more info since the last time I visited the forum and it looked like development was increasingly active, so I jumped to the last page of the thread to see what's happening now. When I got there, I learned two things that really took the wind out of my sails.
The first was that there was a cloud outage that impacted the Bluetti devices, and the company didn't come-clean about it for awhile. I don't know the total impact of the outage, but it was a reminder that relying on the cloud for critical systems like this is a mistake.
More disconcerting was learning that newer Bluetti devices appear to be encrypting the data they send over BLE, making it impossible to monitor or control the devices without using the apps and cloud software that Bluetti provides. I'm sure the justification for this is "security", but given that any phone with the app in range of the device can control it, encrypting the packets doesn't really provide any additional security at that point.
So my idea of making a tool that could be used to integrate these devices with other things like Home Assistant (and ideally, other solar, battery, inverter, etc. devices) in a heterogeneous way is essentially dead due to this.
I'm not particularly frustrated by this because while it's a bummer, it just reminds me of the risks of being dependent on things you have no control over. I'm trying to be more accepting of the fact that I can't do everything myself, but I guess that has to be a dynamic negotiation. Something like backup power isn't really something I can just roll-over and compromise on, especially when it means that it might not work when the "internet is down". Maybe some things are so critical that it's not unreasonable to do them ourselves.
This isn't the first time I've considered designing such things myself and it's certainly something that more and more people are becoming interested in. There may still be an incremental path forward, perhaps making a modification for an existing product? I'm going to explore this option a bit, but I'll keep "doing it from scratch" in my back pocket just in case.
Jason J. Gullickson, 2026